NUMERICAL METHODS

School Of Engineering Of Oils

SYSTEMS OF EQUATIONS
LINEAR AND MATRIXS
M system equations with n unknowns.

It is a set of algebraic expressions of the form:

xj are the unknowns, (j=1,2..., n).

aij are the coefficients, (i=1,2..., m) (j=1,2..., n).

ci are the independent terms, (i=1,2..., m).

The numbers m and n they can be any: m> n, m=n ó m <n.

You will climb them to ij and ci are real numbers.

Climbing aij is the xj coefficient in the i-ésima

equation.

When or n it is small, it is usual to designate to the

unknowns with the letters x, y, z, t...

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

It be observed that the number of equations does not have

why to be equal to the number of unknowns.

When ci=0 for everything i, the system is called

homogeneously.

It is a system of 3 linear equations with 4 unknowns the

coefficients of the first equation of the system are the
numbers 3,-2, 1,-1.

The term independent from the same one is 2.

SOLUTION OF A SYSTEM.

Solution of a system. It is every set of values that it

satisfies to all the equations.
To solve a system. It is to calculate his solution (or
solutions).
Given the system:

1 −4 7
A solution of his is x= ; and¿ ; z=0; t¿ 6 .
6 3

TYPES OF SYSTEMS

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

We can classify the systems attending on the number of

his solutions:

1.Incompatible. It has no solution.

2.Compatible. It has solution.
a. Compatible determined. The only solution.
b. Compatible indeterminate. Infinite solutions.

Incompatible. It has no solution.

Compatible determined. The only

solution.

Compatible indeterminate. Infinite

solutions.

NOTATIONS MATRICIAL

A matrix is a set of elements of any nature although, in

general, there are usually numbers arranged in lines and
columns.

It is called an order matrix "m × n" to a rectangular set

of elements aij arranged in m lines and in n columns. The
order of a matrix also is named a dimension or size,
being a m and n natural numbers.

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

Let's consider a system to be the [1 written in classic

form.

In him there can be considered to be the following

matrixs:

To es the matrix of the coefficients of

order mxn. B is the extended order
matrix mx (n+1).
The system [1] can be written in form
matricial like that:

If in the system [1] we consider the following matrixs:

The system will be written in form vectorial of the

following form:

In this notation, the solutions of a system are the

elements of the form:

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

S = (s1, s2..., sn) Rn

And the following relation happens: To 1S1 + A2S2 +... +

AnSn = C

Let's consider the system:

To es the matrix of the coefficients of order 3x3. B is

the extended order matrix 3x4.

The system can be written of the following forms:

It forms matricial:

It forms vectorial:

In the system:

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

the element S = (-1,1,3,2) is a solution, since it

happens:

TYPES OF MATRIXS
There are some matrixs that appear often and that
according to his form, his elements, receive different
names:

Type of Definition Example

matrix
That matrix that
has only one line,
LINE
being his order
1×n
That matrix that
has only one
COLUMN
column, being his
order m×1
That matrix that
has different
RECTANGUL number of lines
AR that of columns,
being his order
m×n,

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

Considering a
matrix A, it is
called traspuesta
of A to the matrix
that is obtained
TRASPUEST
changing in an
A
organized manner
the lines into the
columns.
It is represented
by Ató AT
The opposite
matrix of the
given one is the
one that is the
result of
PUT UP
replacing every
element with his
opposite one. Put
up of A it is -
to.
If all his
elements are zero.
VOID Also zero is named
a matrix and it is
denoted for 0m×n
SQUARED That matrix that
has equal number
of lines that of
Principal diagonal:
columns, m = n,
they saying that
the matrix
Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

performs order n.

Principal
diagonal: there
are the elements Secondary diagonal:
a11, a22..., ann

Secondary
diagonal: there
are the elements
aij with i+j = n+1

Touch of a square
matrix: there is
the sum of the
elements of the
principal diagonal
a tr A.
It is a square
matrix that is
SYMMETRIC equal to his
AL traspuesta.
A = At, aij = green
pepper
It is a square
matrix that is
equal to put up of
ANTISYMME
his traspuesta.
TRICAL
A =-At, aij = -
green pepper
Necessary aii = 0

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

It is a square
matrix that has
all his void
DIAGONAL
elements except
those of the
principal diagonal
It is a square
matrix that has
all his void
TO CLIMB elements except
those of the
principal diagonal
that are equal
It is a square
matrix that has
all his void
elements except
those of the
IDENTICAL
principal diagonal
that are equal to
1. Also it is
named a matrix
unit.
It is a square
matrix that has
all the elements
TRIANGULA
superficially (for
R
below) of the
principal diagonal
void.

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

An orthogonal
matrix is squared
necessary and
invertible: A-1 = AT

The inverse one of

an orthogonal
matrix is an
ORTHOGONA
orthogonal matrix.
L
The product of two
orthogonal matrixs
is an orthogonal
matrix.
The determinant of
an orthogonal
matrix costs +1 ó-
1.
A matrix is normal
if it exchanges
with his
traspuesta. The
NORMAL symmetrical,
antisymmetrical or
orthogonal matrixs
are necessary
normal.
INVERSE We say that a
square matrix A
has inverse, A-1,
if one verifies
that:
TO · A-1 = A-1 · TO
Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

= I
It is a matrix
that has all his
elements zero,
except those of a
band centred on
STRIPED
the principal
MATRIX
diagonal, for
example a matrix
of this type
tridiagonal will
be:

OPERATIONS BETWEEN MATRIXS

SUM OF MATRIXS

The sum of two matrixs A = (aij) m×nand B = (bij) p×q of

the same dimension (equidimensionales): m = p and n = q
is another matrix C = A+B = (cij) m×n = (aij+bij)

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

It is a law of internal composition with the following

ones

PROPERTIES:

· Associative: To + (B+C) = (A+B) +C

· Commutative: A+B = B+A
· Elem. Neutral: (matrix zero 0m×n), 0+A = A+0 = To
· Elem. Symmetrical: (opposite matrix - to), To + (-to) =
(-to) + To = 0

the sum and difference of two matrixs is not defined if

his dimensions are different!!

PRODUCT OF A REAL NUMBER FOR A MATRIX

To multiply one to climb for a matrix multiplies climbing

for all the elements of the matrix, there being obtained
another matrix of the same order.

It is a law of external composition with the following

ones

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

PROPERTIES:

PRODUCT OF MATRIXS

Given two matrixs A = (aij) m×nand B = (bij) p×q where n =

p, that is to say, the number of columns of the first
matrix A is equal to the number of lines of the matrix B,
the product is defined A · B of the following form:

The element that occupies the place (i, j) in the matrix

product is obtained adding the products of every element
of the line i of the matrix A for the correspondent of
the column j of the matrix B.

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

Associative : To · (B · C) = (To · B) · C

Neutral element: To · I=A

Where I is the matrix identity of the same order

as the matrix A.

It is not Commutative : To · B≠B · To

Distributive of the product with regard to the

sum : To · (B+C) =A · B+A · C

DETERMINANT OF A MATRIX
The determinant of a matrix is a fundamental concept of
the linear algebra with which there decides the existence
and the unit of the results of the systems of linear
equations. To represent the determinant of a matrix A we
will use det. To, although also it gets accustomed to use
the notation [To].

a.If To = [to] es a matrix of 1x1, then det. To = to

b.If To es a matrix of nxn, the minor Mij is the

determinant of the submatrix (n-1) x (n-1) of A, that
is obtained on having suppressed the i-esimo line and
the i-esima column of the matrix A.

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

c.The cofactor associate Aij Mij defines Mij as Aij (-1)

i+j
.

d.The determinant of the matrix to of nxn, when n> 1,

is given for

(Operations investigation: applications and algorithms

Written by Wayne L. Winston)

The determinant is associated with any square matrix A.

To be able to calculate the determinant of a square
matrix, is useful in the study of not linear programming.

Equation 1

Equation 2

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

Definition: if A is a mxm matrix then for any value i and

j, the minor ij-esimo of A (Aij) is the sub matrix (m – 1)
x (m – 1) of A obtained after eliminating the line i and
the column j of A.

For example

Be a mxm matrix. It might represent A like

For cualcular det A, choose any value of i (i=1,2, … …,

m) and determine det A:

Equation 3

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

The previous form names i. an expansion of the det for

way of cofactors of line the advantage of this formula is
that the calculation of the det diminishes A for a matrix
of mxm calculations that need only matrixs (m – 1) x (m –
1). The formula is applied until det A could be expressed
by it in terms of matrixs of 2x2. Later 2 uses the
equation to find the determinants of the matrixs 2x2
pertinent.
In order to illustrate the use of the equation 3 will be
in the det A of

1 expands det A by means of cofactors of the line.

Observe that

a11 = 1, a12 = 2,
a13 = 3.

Also

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

Then for the equation 3

The incomer reader can verify that the det expansion. For
way of the cofactors of the line 2 or of the line 3 also
is equal to zero.

It is a study of the determinants it decides emphasizing

that with their help it is possible to invest square
matrixs and to solve systems of linear equations. But as
is already studied the application of method of Gauss-
Jordán to invest matrixs and to solve systems of linear
equations, already these advantages of the determinants
do not talk each other.

Diego F.
Industrial university De Santander
I Semester of 2010
 NUMERICAL METHODS
School Of Engineering Of Oils

Bibliography
Taken of:

http://www.campusoei.org/cursos/centrocima/matematica/si_
ec_li.pdf
http://personal.redestb.es/ztt/tem/t6_matrices.html

Diego F.
Industrial university De Santander
I Semester of 2010